|
|
|
|
|
""" |
|
|
UltraData Math Parser - Hugging Face Space Demo |
|
|
A unified HTML parser optimized for extracting mathematical content. |
|
|
""" |
|
|
|
|
|
import gradio as gr |
|
|
import requests |
|
|
from ultradata_math_parser import GeneralParser |
|
|
|
|
|
|
|
|
def fetch_url_content(url: str) -> tuple: |
|
|
"""Fetch HTML content from a URL.""" |
|
|
if not url or not url.strip(): |
|
|
return "", "Please enter a URL" |
|
|
|
|
|
url = url.strip() |
|
|
if not url.startswith(("http://", "https://")): |
|
|
url = "https://" + url |
|
|
|
|
|
try: |
|
|
headers = { |
|
|
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36" |
|
|
} |
|
|
response = requests.get(url, headers=headers, timeout=15) |
|
|
response.raise_for_status() |
|
|
return response.text, url |
|
|
except requests.exceptions.Timeout: |
|
|
return "", f"Request timed out for {url}" |
|
|
except requests.exceptions.RequestException as e: |
|
|
return "", f"Failed to fetch URL: {str(e)}" |
|
|
|
|
|
|
|
|
def fetch_and_parse(url: str, process_math: bool, include_tables: bool, enable_forum: bool, html_type: str) -> tuple: |
|
|
"""Fetch URL content and parse it in one step.""" |
|
|
html_content, base_url = fetch_url_content(url) |
|
|
|
|
|
if not html_content: |
|
|
|
|
|
error_msg = base_url |
|
|
return "", error_msg, f"❌ {error_msg}", "", "", f"**Error:** {error_msg}" |
|
|
|
|
|
result = parse_html( |
|
|
html_content=html_content, |
|
|
base_url=base_url, |
|
|
process_math=process_math, |
|
|
include_tables=include_tables, |
|
|
enable_forum_assembly=enable_forum, |
|
|
html_type=html_type, |
|
|
) |
|
|
|
|
|
formatted = format_output(result) |
|
|
|
|
|
return html_content, base_url, formatted[0], formatted[1], formatted[2], formatted[3] |
|
|
|
|
|
|
|
|
def parse_html( |
|
|
html_content: str, |
|
|
base_url: str = "", |
|
|
process_math: bool = True, |
|
|
include_tables: bool = True, |
|
|
enable_forum_assembly: bool = True, |
|
|
html_type: str = "unified", |
|
|
) -> dict: |
|
|
""" |
|
|
Parse HTML content using GeneralParser. |
|
|
|
|
|
Args: |
|
|
html_content: Raw HTML string to parse |
|
|
base_url: Base URL for resolving relative links |
|
|
process_math: Whether to process and convert math expressions |
|
|
include_tables: Whether to preserve table elements |
|
|
enable_forum_assembly: Whether to enable forum post assembly |
|
|
html_type: Parser type (unified/article/forum) |
|
|
|
|
|
Returns: |
|
|
Dictionary containing parsed results |
|
|
""" |
|
|
if not html_content or not html_content.strip(): |
|
|
return { |
|
|
"title": "", |
|
|
"html": "", |
|
|
"text": "", |
|
|
"text_length": 0, |
|
|
"xp_num": "", |
|
|
"fallback_strategy": "", |
|
|
"forum_assembled": False, |
|
|
"error": "Please provide HTML content to parse.", |
|
|
} |
|
|
|
|
|
parser = GeneralParser() |
|
|
|
|
|
try: |
|
|
result = parser.extract( |
|
|
html=html_content, |
|
|
base_url=base_url, |
|
|
process_math=process_math, |
|
|
include_tables=include_tables, |
|
|
enable_forum_assembly=enable_forum_assembly, |
|
|
html_type=html_type, |
|
|
) |
|
|
|
|
|
return { |
|
|
"title": result.get("title", ""), |
|
|
"html": result.get("html", ""), |
|
|
"text": result.get("text", ""), |
|
|
"text_length": result.get("text_length", 0), |
|
|
"xp_num": result.get("xp_num", ""), |
|
|
"fallback_strategy": result.get("fallback_strategy", ""), |
|
|
"forum_assembled": result.get("forum_assembled", False), |
|
|
"error": None, |
|
|
} |
|
|
except Exception as e: |
|
|
return { |
|
|
"title": "", |
|
|
"html": "", |
|
|
"text": "", |
|
|
"text_length": 0, |
|
|
"xp_num": "", |
|
|
"fallback_strategy": "", |
|
|
"forum_assembled": False, |
|
|
"error": str(e), |
|
|
} |
|
|
|
|
|
|
|
|
def format_output(result: dict) -> tuple: |
|
|
"""Format the parser output for Gradio display.""" |
|
|
if result.get("error"): |
|
|
return ( |
|
|
f"❌ Error: {result['error']}", |
|
|
"", |
|
|
"", |
|
|
f"**Error:** {result['error']}", |
|
|
) |
|
|
|
|
|
|
|
|
text_content = result.get("text", "") |
|
|
markdown_content = text_content if text_content else "_No content extracted_" |
|
|
|
|
|
return ( |
|
|
result.get("title", ""), |
|
|
result.get("html", ""), |
|
|
result.get("text", ""), |
|
|
markdown_content, |
|
|
) |
|
|
|
|
|
|
|
|
def process_input(html_content, base_url, process_math, include_tables, enable_forum, html_type): |
|
|
"""Main processing function for Gradio interface.""" |
|
|
result = parse_html( |
|
|
html_content=html_content, |
|
|
base_url=base_url, |
|
|
process_math=process_math, |
|
|
include_tables=include_tables, |
|
|
enable_forum_assembly=enable_forum, |
|
|
html_type=html_type, |
|
|
) |
|
|
return format_output(result) |
|
|
|
|
|
|
|
|
|
|
|
EXAMPLE_HTML = """<!DOCTYPE html> |
|
|
<html> |
|
|
<head> |
|
|
<title>Quadratic Formula Example</title> |
|
|
</head> |
|
|
<body> |
|
|
<article class="post-content"> |
|
|
<h1>Understanding the Quadratic Formula</h1> |
|
|
<p>The quadratic formula is used to solve equations of the form ax² + bx + c = 0.</p> |
|
|
<p>The solution is given by:</p> |
|
|
<math xmlns="http://www.w3.org/1998/Math/MathML"> |
|
|
<mi>x</mi> |
|
|
<mo>=</mo> |
|
|
<mfrac> |
|
|
<mrow> |
|
|
<mo>-</mo> |
|
|
<mi>b</mi> |
|
|
<mo>±</mo> |
|
|
<msqrt> |
|
|
<mrow> |
|
|
<msup><mi>b</mi><mn>2</mn></msup> |
|
|
<mo>-</mo> |
|
|
<mn>4</mn> |
|
|
<mi>a</mi> |
|
|
<mi>c</mi> |
|
|
</mrow> |
|
|
</msqrt> |
|
|
</mrow> |
|
|
<mrow> |
|
|
<mn>2</mn> |
|
|
<mi>a</mi> |
|
|
</mrow> |
|
|
</mfrac> |
|
|
</math> |
|
|
<p>Where a, b, and c are coefficients of the quadratic equation.</p> |
|
|
<h2>Example Problem</h2> |
|
|
<p>Solve: x² - 5x + 6 = 0</p> |
|
|
<p>Here, a = 1, b = -5, c = 6</p> |
|
|
<p>Using the formula: x = (5 ± √(25-24))/2 = (5 ± 1)/2</p> |
|
|
<p>Therefore, x = 3 or x = 2</p> |
|
|
</article> |
|
|
<footer> |
|
|
<nav>Related articles...</nav> |
|
|
</footer> |
|
|
</body> |
|
|
</html>""" |
|
|
|
|
|
|
|
|
|
|
|
custom_css = """ |
|
|
@import url('https://fonts.googleapis.com/css2?family=Inter:wght@300;400;600;700&family=JetBrains+Mono:wght@400;500&display=swap'); |
|
|
|
|
|
:root { |
|
|
--bg: #f3f5fb; |
|
|
--card: #ffffff; |
|
|
--border: #e5e9f2; |
|
|
--text: #1f2a44; |
|
|
--muted: #6b7280; |
|
|
--accent: #5b6ff4; |
|
|
--accent-2: #7a8af7; |
|
|
--input: #f7f9fc; |
|
|
} |
|
|
|
|
|
body { |
|
|
background-color: var(--bg) !important; |
|
|
color: var(--text) !important; |
|
|
} |
|
|
|
|
|
.gradio-container { |
|
|
font-family: 'Inter', sans-serif !important; |
|
|
background: var(--bg) !important; |
|
|
width: 100% !important; |
|
|
max-width: none !important; |
|
|
margin: 0 auto !important; |
|
|
padding: clamp(12px, 2vw, 28px) clamp(10px, 3vw, 32px) clamp(20px, 3vw, 48px) !important; |
|
|
} |
|
|
|
|
|
.main-title { |
|
|
text-align: center !important; |
|
|
font-weight: 800 !important; |
|
|
font-size: 2.4rem !important; |
|
|
color: #3b3fb6 !important; |
|
|
margin-bottom: 0.25rem !important; |
|
|
} |
|
|
|
|
|
.subtitle { |
|
|
text-align: center !important; |
|
|
color: var(--muted) !important; |
|
|
font-size: 1rem !important; |
|
|
margin-bottom: 1.5rem !important; |
|
|
} |
|
|
|
|
|
.glass-panel { |
|
|
background: var(--card) !important; |
|
|
border: 1px solid var(--border) !important; |
|
|
border-radius: 16px !important; |
|
|
padding: 20px !important; |
|
|
box-shadow: 0 10px 24px rgba(31, 42, 68, 0.08) !important; |
|
|
} |
|
|
|
|
|
.section-header { |
|
|
color: var(--text) !important; |
|
|
font-weight: 700 !important; |
|
|
font-size: 1.05rem !important; |
|
|
margin-bottom: 1rem !important; |
|
|
padding-bottom: 0.5rem !important; |
|
|
border-bottom: 1px solid var(--border) !important; |
|
|
} |
|
|
|
|
|
.block > label > span, |
|
|
.form > label > span, |
|
|
.gr-form > label > span, |
|
|
.label-wrap > span, |
|
|
span.description, |
|
|
.description { |
|
|
color: var(--text) !important; |
|
|
text-shadow: none !important; |
|
|
} |
|
|
|
|
|
fieldset legend, fieldset legend span, |
|
|
.gr-radio > label, .gr-radio > label span { |
|
|
color: var(--text) !important; |
|
|
font-weight: 600 !important; |
|
|
} |
|
|
|
|
|
.gr-input, textarea, input, .gr-box, .gr-check-radio, .gr-dropdown { |
|
|
font-family: 'JetBrains Mono', monospace !important; |
|
|
background-color: var(--input) !important; |
|
|
border: 1px solid var(--border) !important; |
|
|
color: var(--text) !important; |
|
|
box-shadow: none !important; |
|
|
} |
|
|
|
|
|
.gr-input:focus, textarea:focus, input:focus { |
|
|
border-color: var(--accent) !important; |
|
|
background-color: #ffffff !important; |
|
|
} |
|
|
|
|
|
.tabs { |
|
|
border: none !important; |
|
|
margin-bottom: 1rem !important; |
|
|
} |
|
|
|
|
|
.tab-nav { |
|
|
border-bottom: 1px solid var(--border) !important; |
|
|
justify-content: center !important; |
|
|
gap: 8px !important; |
|
|
} |
|
|
|
|
|
.tab-nav button { |
|
|
font-weight: 600 !important; |
|
|
font-size: 0.95rem !important; |
|
|
color: var(--muted) !important; |
|
|
background: transparent !important; |
|
|
border-radius: 8px !important; |
|
|
padding: 6px 12px !important; |
|
|
} |
|
|
|
|
|
.tab-nav button.selected { |
|
|
color: #ffffff !important; |
|
|
background: var(--accent) !important; |
|
|
border-bottom: none !important; |
|
|
} |
|
|
|
|
|
.gr-button-primary { |
|
|
background: linear-gradient(135deg, var(--accent) 0%, var(--accent-2) 100%) !important; |
|
|
border: none !important; |
|
|
color: #ffffff !important; |
|
|
font-weight: 600 !important; |
|
|
border-radius: 8px !important; |
|
|
box-shadow: 0 6px 14px rgba(91, 111, 244, 0.28) !important; |
|
|
} |
|
|
|
|
|
.gr-button-primary:hover { |
|
|
filter: brightness(0.98) !important; |
|
|
} |
|
|
|
|
|
.gr-button-secondary { |
|
|
background: #f1f4fb !important; |
|
|
border: 1px solid var(--border) !important; |
|
|
color: var(--text) !important; |
|
|
border-radius: 8px !important; |
|
|
} |
|
|
|
|
|
.output-textbox, .markdown-box { |
|
|
background-color: #fbfcff !important; |
|
|
border-radius: 12px !important; |
|
|
} |
|
|
|
|
|
.output-textbox textarea { |
|
|
background-color: transparent !important; |
|
|
border: none !important; |
|
|
box-shadow: none !important; |
|
|
color: var(--text) !important; |
|
|
} |
|
|
|
|
|
.markdown-box .prose { |
|
|
color: var(--text) !important; |
|
|
border: none !important; |
|
|
box-shadow: none !important; |
|
|
background: transparent !important; |
|
|
} |
|
|
|
|
|
.markdown-box .prose, |
|
|
.markdown-box .prose * , |
|
|
.markdown-box .md, |
|
|
.markdown-box .wrap, |
|
|
.markdown-box .wrap * , |
|
|
.markdown-box .label-wrap, |
|
|
.markdown-box .block, |
|
|
.markdown-box .form, |
|
|
.markdown-box .container, |
|
|
.markdown-box .box { |
|
|
border: none !important; |
|
|
box-shadow: none !important; |
|
|
background: transparent !important; |
|
|
padding: 0 !important; |
|
|
} |
|
|
|
|
|
/* Target the specific container that usually holds the markdown output */ |
|
|
.markdown-box > .wrap > .block, |
|
|
.markdown-box > .block { |
|
|
border: none !important; |
|
|
background: transparent !important; |
|
|
} |
|
|
|
|
|
.footer-text { |
|
|
text-align: center !important; |
|
|
margin-top: 2rem !important; |
|
|
padding-top: 1.5rem !important; |
|
|
color: var(--muted) !important; |
|
|
border-top: 1px solid var(--border) !important; |
|
|
font-size: 0.9rem !important; |
|
|
} |
|
|
""" |
|
|
|
|
|
|
|
|
with gr.Blocks(title="UltraData-Math-L0-Parser", css=custom_css, theme=gr.themes.Soft()) as demo: |
|
|
gr.HTML('<h1 class="main-title">UltraData-Math-L0-Parser</h1>') |
|
|
gr.HTML('<p class="subtitle">Unified HTML Parser for Mathematical Content Extraction</p>') |
|
|
|
|
|
with gr.Row(): |
|
|
with gr.Column(scale=1, elem_classes=["glass-panel"]): |
|
|
gr.HTML('<div class="section-header">Input</div>') |
|
|
|
|
|
with gr.Tabs(): |
|
|
with gr.TabItem("🔗 URL"): |
|
|
url_input = gr.Textbox( |
|
|
label="URL", |
|
|
placeholder="Enter URL to fetch (e.g., https://example.com/math-article)", |
|
|
lines=3, |
|
|
max_lines=5, |
|
|
value="https://math.stackexchange.com/questions/5120625/ode-problem-of-yt-sqrtyt-with-the-inital-value-y0-1-t-geq-0", |
|
|
) |
|
|
fetch_btn = gr.Button("Fetch & Parse", variant="primary", size="lg") |
|
|
|
|
|
with gr.TabItem("📝 HTML"): |
|
|
pass |
|
|
|
|
|
html_input = gr.Textbox( |
|
|
label="HTML Content", |
|
|
placeholder="Paste your HTML content here or fetch from URL above...", |
|
|
lines=10, |
|
|
max_lines=20, |
|
|
value=EXAMPLE_HTML, |
|
|
) |
|
|
|
|
|
base_url_input = gr.Textbox( |
|
|
label="Base URL (Auto-filled from URL fetch)", |
|
|
placeholder="https://example.com/page", |
|
|
lines=1, |
|
|
) |
|
|
|
|
|
with gr.Accordion("⚙️ Advanced Options", open=False, visible=False): |
|
|
html_type = gr.Radio( |
|
|
choices=["unified", "article", "forum"], |
|
|
value="unified", |
|
|
label="Parser Type", |
|
|
info="Select the parsing strategy", |
|
|
) |
|
|
process_math = gr.Checkbox( |
|
|
label="Process Math Expressions", |
|
|
value=True, |
|
|
info="Convert MathML and LaTeX to unified format", |
|
|
) |
|
|
include_tables = gr.Checkbox( |
|
|
label="Include Tables", |
|
|
value=True, |
|
|
info="Preserve table elements in output", |
|
|
) |
|
|
enable_forum = gr.Checkbox( |
|
|
label="Enable Forum Assembly", |
|
|
value=True, |
|
|
info="Assemble forum posts and comments", |
|
|
) |
|
|
|
|
|
with gr.Row(): |
|
|
parse_btn = gr.Button("🚀 Parse HTML", variant="primary", size="lg") |
|
|
clear_btn = gr.Button("🗑️ Clear", variant="secondary", size="lg") |
|
|
|
|
|
with gr.Column(scale=1, elem_classes=["glass-panel"]): |
|
|
gr.HTML('<div class="section-header">Output</div>') |
|
|
|
|
|
title_output = gr.Textbox( |
|
|
label="Extracted Title", |
|
|
lines=1, |
|
|
interactive=False, |
|
|
) |
|
|
|
|
|
with gr.Tabs(): |
|
|
with gr.TabItem("✨ Markdown"): |
|
|
markdown_output = gr.Markdown( |
|
|
value="### Output will appear here...", |
|
|
label="Markdown Preview", |
|
|
elem_classes=["markdown-box"], |
|
|
latex_delimiters=[ |
|
|
{"left": "$$", "right": "$$", "display": True}, |
|
|
{"left": "$", "right": "$", "display": False}, |
|
|
{"left": "\\[", "right": "\\]", "display": True}, |
|
|
{"left": "\\(", "right": "\\)", "display": False}, |
|
|
], |
|
|
) |
|
|
with gr.TabItem("📄 Plain Text"): |
|
|
text_output = gr.Textbox( |
|
|
label="Plain Text (w3m rendered)", |
|
|
lines=25, |
|
|
max_lines=30, |
|
|
interactive=False, |
|
|
autoscroll=False, |
|
|
elem_classes=["output-textbox"], |
|
|
) |
|
|
with gr.TabItem("📝 Raw HTML"): |
|
|
html_output = gr.Textbox( |
|
|
label="Extracted HTML", |
|
|
lines=25, |
|
|
max_lines=30, |
|
|
interactive=False, |
|
|
autoscroll=False, |
|
|
elem_classes=["output-textbox"], |
|
|
) |
|
|
|
|
|
|
|
|
fetch_btn.click( |
|
|
fn=fetch_and_parse, |
|
|
inputs=[url_input, process_math, include_tables, enable_forum, html_type], |
|
|
outputs=[html_input, base_url_input, title_output, html_output, text_output, markdown_output], |
|
|
) |
|
|
|
|
|
parse_btn.click( |
|
|
fn=process_input, |
|
|
inputs=[html_input, base_url_input, process_math, include_tables, enable_forum, html_type], |
|
|
outputs=[title_output, html_output, text_output, markdown_output], |
|
|
) |
|
|
|
|
|
def clear_all(): |
|
|
return "", "", "", "", "", "", "" |
|
|
|
|
|
clear_btn.click( |
|
|
fn=clear_all, |
|
|
outputs=[url_input, html_input, base_url_input, title_output, html_output, text_output, markdown_output], |
|
|
) |
|
|
|
|
|
|
|
|
gr.HTML(""" |
|
|
<div class="footer-text"> |
|
|
<p>🔬 <strong>UltraData Math Parser</strong> - Part of the UltraData-Math Project</p> |
|
|
<p>Specialized in extracting mathematical content from web pages with MathML, LaTeX, and formula support.</p> |
|
|
</div> |
|
|
""") |
|
|
|
|
|
|
|
|
if __name__ == "__main__": |
|
|
demo.launch(ssr_mode=False) |
|
|
|
